People living in southern China had the most unique gut signatures. Credit: ANATOMICAL TRAVELOGUE/SPL

Thousands of frozen urine samples have yielded new information about the diversity of human metabolism across the globe — about who eats what, and how their unique internal microorganisms handle the input.

‘Genome-wide association studies’ can link specific gene variants with diseases and predictors of disease, such as blood pressure and weight. But most diseases are caused by a complex interaction between genes, lifestyle factors such as diet and, as is increasingly being recognized, the trillions of microbes that reside in the human gut. To get information about those interactions, a ‘metabolome-wide association study’ looks at the products people excrete1.

A team led by Elaine Holmes and Ruey Leng Loo of Imperial College London took advantage of an older epidemiological study on diet and blood pressure that collected urine samples from 4,680 people between 1997 and 1999. These samples were analysed, and the results published in 20032, then preserved with boric acid and kept frozen.

The research team were able to do with most of the samples something not possible in the original study: identify all the chemical compounds in the urine, using an analytical technique called proton nuclear magnetic resonance spectroscopy.

From the gut

The method produces a graph with thousands of peaks, each of which corresponds to a different metabolite, the compounds left over after the body is done digesting food. The researchers then compared these graphs across the 17 populations of subjects, who came from China, Japan, the United Kingdom and the United States. "Of the thousands of peaks, we find the 20, 30 or 40 that are different" from each other, says team member Jeremy Nicholson, also from Imperial College.

With this method, they can not only link specific metabolites to specific diseases, they can also compare the entire suite of metabolites — the metabolome — between groups.

"What our study really shows is how incredibly metabolically diverse people are around the world," says Nicholson. "British and American [metabolomes] are nearly identical. Japanese and Chinese people are totally different metabolically even though they are nearly identical genetically." People who lived in Hawaii had metabolomes equally similar to those of people on the mainland United States and in Japan.

Interestingly, Nicholson says, the biggest difference between the 17 groups was between people from South China and everyone else. "They have a very different and much broader range of diet," he says. "Very broadly speaking, the southern Chinese are the healthiest and the people in southern Texas are least healthy."

The group also found some intriguing relationships between high blood pressure and several metabolites, including formate and hippurate. Formate is produced in several ways in the body, some undertaken by gut microbes and some undertaken by the kidney. It is already understood to be involved with the kidney's chloride balance, which is linked to blood pressure, so the fact that people with high formate levels had high blood pressure makes sense. But for hippurate, the story is less clear. People with less hippurate had higher blood pressure, and more work is needed to understand why.

Next steps

Teri Manolio, who runs the office of population genomics at the National Human Genome Research Institute in Bethesda, Maryland, says that the study finally demonstrates the feasibility of screening many metabolomes at one time. "It opens up a whole new avenue to looking at diet and microbes between populations," she says. "This could be the paradigm-setting study where people say that this is feasible and reproducible."

Manolio notes that the effect of gut microbes on health has been underestimated. "It is what is in the gut and elsewhere that influences the way we process foods and respond to infection," she says.